/
DESCRIPTION
34 lines (33 loc) · 1.38 KB
/
DESCRIPTION
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Package: bouldr
Title: A Psychology-Focused Tool for ROC Analyses
Version: 0.1.2
Authors@R: c(
person(given = "Joshua",
family = "Langfus",
role = c("aut", "cre"),
email = "me@josh-langfus.com",
comment = c(ORCID = "0000-0003-1096-966X")),
person(given = "Sandra",
family = "Williams",
role = "ctb")
)
Author: Joshua Langfus [aut, cre]
Maintainer: Joshua Langfus <me@josh-langfus.com>
Description: Extends the pROC package to do multiple nested Receiver Operator
Characteristic (ROC) curve analyses, and graph them nicely with ggplot2.
Many ROC packages focus on machine learning and classification use-cases.
However, ROC is useful in psychology as well -- particularly in developing
assessment instruments. A continuous score on a test can be ROC-ed against
a target diagnosis to test the diagnostic efficiency of the test. ROC is
the preferred approach because it assesses sensativity and specificity
across the range of the scale, and can be used to identify optimal cut-
scores. This package relies on pROC to perform the computation.
Depends: R (>= 3.5.0)
Imports: pROC, tidyr, dplyr, rlang, magrittr, RcppAlgos, broom, ggplot2, stats
Suggests: knitr, rmarkdown
VignetteBuilder: knitr
License: file LICENSE
Encoding: UTF-8
LazyData: true
Roxygen: list(markdown = TRUE)
RoxygenNote: 7.3.1